The 7th Financial Narrative Processing Workshop (FNP 2026)
CALL FOR PAPERS
We are delighted to announce the 7th Financial Narrative Processing Workshop (FNP 2026), an international event dedicated to the intersection of Natural Language Processing (NLP), Machine Learning (ML), and Financial/Accounting data.
This year, FNP 2026 will be held in Palma de Mallorca, Spain, within the framework of LREC 2026. The workshop will run as a one-day hybrid event, allowing for both on-site interaction and online presentation options for authors unable to travel. FNP 2026 marks
our 11th international event in Financial NLP and the 7th dedicated FNP workshop. Our community has consistently bridged the gap between computational linguists, data scientists, economists, and accounting scholars. Previous editions have demonstrated strong
scholarly impact, with proceedings indexed in the ACL Anthology and IEEE Xplore, and over 100 shared task papers published.
Invited speaker:
Dr Pablo Haya
Head of Business and Language Analytics (BLA), IIC – Universidad Autónoma de Madrid
Call for Papers (main track):
We invite submissions of original, unpublished research in Financial NLP and Financial Text Analysis. We welcome work covering theoretical insights, methodological innovations, datasets, resources, applied systems, and reflections on practical challenges. Negative
results are also welcome.
Topics include (but are not limited to):
-
Core technologies for financial narratives: morphological analysis, tokenisation, disambiguation, part-of-speech tagging, named entity recognition, chunking, parsing, semantic role labelling, sentiment analysis, document quality, and
advanced readability metrics.
-
Ethics, diversity, and wellbeing: using NLP to detect misreporting and biases in financial narratives relating to gender, ethnicity, representation of women at work, employee wellbeing, mental health, and organisational stability.
-
Resources and tools: financial dictionaries, annotated corpora, benchmarks, software, and novel methodologies.
-
Summarisation in financial contexts: single- and multi-document summarisation, headline generation, evaluation of summaries, and cross-domain or cross-lingual summarisation (e.g., company blogs, market briefs, product reviews).
-
Social media and financial discourse: analysing online platforms to capture public opinion and sentiment around financial events.
-
Multilingual and cross-regulatory perspectives: exploring financial narratives across languages, dialects, and varying international regulatory regimes.
-
Ongoing and preliminary research: work-in-progress studies with promising early findings.
-
Negative results: experiments showing where methods (including state-of-the-art models such as BERT and its variants) succeed or fail across languages, domains, or tasks.
(Note: In addition to the Main Track, FNP 2026 will also host the FinCausal 2026 shared task. A separate call will be issued for this competition.)
Submission Types:
Formatting will follow the LREC 2026 guidelines.
Long Papers: Up to 8 pages. Should present substantial, original, completed, and unpublished work.
Short Papers: Up to 4 pages. Suitable for describing small focused contributions, negative results, or system demonstrations.
Papers MUST be submitted via the conference submission system (link will be available on the workshop website). All submissions will undergo a double-blind peer review process.
Key Dates:
4 Dec 2025 – First Call for Papers
7 Jan 2026 – Shared-task training set release
10 Jan 2026 – Second Call for Papers
1 Feb 2026 – Shared-task blind test release
15 Feb 2026 – Shared-task system submissions
20 Feb 2026 – Paper submission deadline
11 Mar 2026 – Acceptance notifications
30 Mar 2026 – Camera-ready deadline
11–16 May 2026 – LREC 2026 & Workshops (final date TBC)
Organising Committee:
Shared Task Chair — Dr Jordi Porta (UAM / Real Academia Española, Spain) —
jordi...@uam.es
Scholarly Impact:
FNP has built a strong research presence, producing influential datasets, widely cited papers, and long-running shared tasks. Citation records are publicly available: